gaussdb-ro-mcp
Server Configuration
Describes the environment variables required to run the server.
| Name | Required | Description | Default |
|---|---|---|---|
| GAUSSDB_RO_MCP_CONFIG | No | Path to the GaussDB read-only MCP server's YAML configuration file. Defaults to './gaussdb-ro-mcp.yaml' if not set. | ./gaussdb-ro-mcp.yaml |
Instructions
Guidance the server publishes about itself, which clients place ahead of the tool catalog so the model reads it before choosing anything.
This server publishes no instructions, or was last inspected before Glama recorded them.
Capabilities
Features and capabilities supported by this server
Protocol revision2025-11-25
| Capability | Details |
|---|---|
| tools | {
"listChanged": true
} |
| logging | {} |
Tools
Functions exposed to the LLM to take actions
| Name | Description |
|---|---|
| describe_tableA | 查看表/视图的完整结构:列清单(类型/可空/默认值/注释)、主键与约束、全部索引及其定义;视图返回视图定义 SQL,分区表返回分区清单。适用场景:编写或校对查询前了解表结构、关联键与索引设计。table 为必填(可来自 list_tables 的结果);schema 省略时自动在非系统模式中查找,存在同名表时请显式指定 schema。 |
| execute_selectA | 执行只读 SELECT 查询(本服务器唯一允许的 SQL 入口)。仅接受单条 SELECT/WITH 语句:拒绝 DML/DDL、SELECT INTO、FOR UPDATE/SHARE 行锁、多语句与危险函数;查询在服务端强制的只读事务中执行。适用场景:在 list_tables/describe_table 定位目标后做数据探查与统计(计数、聚合、抽样、排序取前 N 行)。max_rows 控制本次返回行数(默认取实例配置,受服务端硬上限约束);大结果集建议用 LIMIT 或聚合减少传输。 |
| list_schemasA | 列出 GaussDB 数据库中的 schema(模式)清单,含每个模式下的对象数量与注释,默认排除系统模式。适用场景:探查数据库的第一步——先确定业务模式名称,再用 list_tables 查看其中的表。include_system 为 true 时包含 pg_catalog、information_schema 等系统模式(一般仅在排查系统对象时需要)。 |
| list_tablesA | 列出 GaussDB 数据库中的表和视图清单:名称、类型(表/视图/物化视图/分区表/外表)、估算行数与注释。schema 参数可过滤到指定模式(可来自 list_schemas 的结果),省略时扫描全部非系统模式。适用场景:定位目标表后,用 describe_table 查看结构或用 execute_select 抽样查询。大型库建议传 schema 缩小扫描范围。 |
| test_connectionA | 测试 GaussDB 数据源连通性,返回服务器版本、当前数据库、用户、只读状态与连接耗时。建议在会话开始或其他工具调用前先执行,确认数据源可达且只读事务生效;也用于排查连接、凭据或超时问题。不指定 instance 时测试默认实例。 |
Prompts
Interactive templates invoked by user choice
| Name | Description |
|---|---|
No prompts | |
Resources
Contextual data attached and managed by the client
| Name | Description |
|---|---|
No resources | |
TDQS
Scored across 5 tools
Each tool has a clearly distinct role: connectivity testing, schema listing, table listing, table structure inspection, and read-only query execution. The discovery flow is explicit and there is no overlap between tools.
All tool names use consistent snake_case verb_noun naming: list_schemas, list_tables, describe_table, execute_select, test_connection. The convention is predictable throughout.
Five tools is well-scoped for a read-only GaussDB exploration server. Each tool earns its place and supports a distinct step in the discovery/query workflow.
The surface covers the full read-only lifecycle: connectivity check, schema discovery, table/view discovery, structural inspection, and SELECT querying. No write operations are expected for a read-only server, and no obvious dead ends are present.